Exploitation of knowledge databases in the synthesis of zinc(II) malonates with photo-sensitive and photo-insensitive N,N′-containing linkers

Crystallographic analysis of zinc(II) complexes allows the construction of hydrogen-bonded and coordination networks with 1,2-bis(pyridin-4-yl)ethylene groups situated in photoreactive positions to allow solid-state [2 + 2] cycloaddition reactions.

Ratio -ratio of L and Zn II ; CN -coordination number of Zn II ; CP -coordination polyhedron; CF -coordination formula; possible nets are taken from previously reported databased of networks obtained for coordination polymers [S1- S3] or taken for a database of zinc(II) complexes with bipy, bpe or bpa.  2

S1.1. Coordination Formulas and Their Applications
Let us denote mono-, bi-, tri-or tetradenate ligands with M, B, T or K letters. The way in which metal atoms A surround the ligand is denoted by numerical superscripts (mbtk). The superscripts define the 'partial'denticity of the ligand with respect to any A atom (mmono-, bbi-, ttri-, ktetradenticity). The number of A atoms with respect to the ligand that exhibits the corresponding partial denticity is denoted by the numerical value of the corresponding superscript. Then the coordination type of an i-th ligand is given as D mbtk i . A few examples of tridentate ligands coordinated by one, two or three metal atoms are given in Fig. S1, as well as the corresponding coordination-type symbols.

Figure S1
Selected coordination modes of bpe or malonate anion and its derivatives in zinc(II) complexes.
The symbol for the ligand coordination type also denotes the total number of complexing atoms (Z) which surround the ligand, and the total number of chemical bonds that the ligand makes with the central atom (N B ) as Z = m + b + t + k and N B = 1m + 2b + 3t + 4k.
Provided that the coordination types of all the ligands in a complex are determined, the coordination formula (CF) of the complex can be written. Any CF includes the coordination types of all the ligands with the same chemical formula (with the exception of counterions and molecules). The subscripts denote the stoichiometric composition with respect to any equivalent ligand and a metal A atom. Using the chemical and crystallochemical formulae of a complex together allows the environment of the central atom to be characterized in order to calculate the coordination number (CN) and the number of ligands in the first coordination sphere (N A ) without any diagramatical or text description.

S1.2. Analysis of Previously Reported Zinc(II) Complexes with bipy, bpe and bpa.
Distribution of zinc(II) coordination numbers and composition of coordination polyhedra has been calculated for 1473 complexes containing both zinc(II) and at least one of bipy, bpe or bpa ligands. Distribution of zinc(II) coordination numbers is given on Fig. S2; and distribution of various coordination polyhedra for the most widespread coordination numbers 4 -6 is depicted on Fig. S3. Note, that the number of coordination polyhedra O 4 , O 5 and O 6 is substantially non-zero, but these were excluded from analysis as we were interested only in mixed complexes containing L ligands. The number of polyhedra containing three and more nitrogen atoms is also high, but these can appear only if L : Zn II = 2 : 1 and more.

Figure S2
Distribution of coordination numbers for zinc(II) atoms in Zn II N x O y coordination polyhedra found in 1473 X-rayed complexes with bipy, bpe or bpa ligands.   Aldrich», Germany, 97%). IR spectra were measured by using a Perkin-Elmer Spectrum 65 instrument by the attenuated total reflection (ATR) method in the range 4000-400 cm -1 . CHN analysis was performed by using an automatic CHNS analyzer EuroEA3000 at the Center of Collective Use of IGIC RAS.

S1.4. Crystallography
Single crystals of 1-9 were obtained from reaction mixtures. The intensities of reflections were measured with a Bruker Apex II DUO CCD diffractometer using graphite monochromated MoK  radiation ( = 0.71073 Å) at 120.0(2) K. Intensity data for 2a were collected at the at the K4.4 station of the Kurchatov Center for Synchrotron Radiation and Nanotechnology in Moscow (Russia) at a wavelength of 0.9699 Å using a MAR CCD 165 detector and merged using SCALA. [S4] Data collection was performed at low temperature [100 K] using an Oxford CryoJet from Oxford Cryosystems Ltd. The structures were solved by the direct method and refined by full-matrix least squares against F 2 . Non-hydrogen atoms were refined anisotropically except some disordered atoms. The disordered fragments, particularly, one carbon atom of bipy in 1, solvent bpe molecule in 7, one ethyl fragment in 9, a methyl group and all carbon atoms of bpe and tpcb ligands in 2a were refined isotropically. A number of EADP, ISOR, SADI, RIGU and DFIX instructions were applied to refine some moieties, especially, in crystals of 2a, 7, 8a disordered by symmetry or containing disordered fragments.
TWIN/BASF refinement was performed for 9. Positions of hydrogen atoms were calculated and all were included in the refinement by the riding model with U iso (H) = 1.5U eq (X) for methyl groups and water molecules, and U iso (H) = 1.2U eq (X) for other atoms. All calculations were made using the SHELXL2014 [S5] and OLEX2 [S6] program packages. Experimental details and crystal parameters are listed in Tables S4 and S5.

S1.5. Powder X-Ray diffraction.
Phase composition of the bulk samples was confirmed with powder XRD. Powder patterns were measured on a Bruker D8 Advance diffractometer at room temperature with LynxEye detector and Ge(111) monochromator, λ(CuKα 1 ) = 1.54060 Å, θ/2θ scan from 4° to 60°. The powder patterns were modeled with the Rietveld method using Bruker TOPAS5 [S7] software. Fundamental parameters approach (Cheary & Coelho, 1992) was used for profile fitting. The preferred orientation was taken into account with the spherical harmonics approach (Järvinen, 1993).

Figure S10
The experimental (blue) and calculated (red) powder patterns for [Zn(bpe)(Et 2 mal)] (7) and their difference (grey). Smooth residual curve and R wp /R bragg = 17.73/3.88 % indicate that the sample exhibit strong prefered orientation and consists mainly from the target substance. Some impurity is present that we failed to determine.

Figure S11
The experimental (blue) and calculated (red) powder patterns for [Zn(bpe)(Et 2 mal)] (7) irradiated for 6 hours. Rietveld analysis indicates purity of the sample. R wp = 5.296% and R bragg = 1.030%. The blue line is the experimental pattern, the fuchsia line is the calculated pattern, and the grey line is the difference curve.

Figure S12
The experimental (blue) and calculated (red) powder patterns for [Zn(H 2 O) 4 (bpe) 2 ](HEt 2 mal) 2 (8) and their difference (grey). Rietveld analysis indicates that the sample consists mainly from the target substance. Some impurity is present that we failed to determine.

Figure S13
The experimental (blue) and calculated (red) powder patterns for